How Many Studies Does it Take to Change a Lightbulb? Part 1

By Noah Sabatier

Changing a light bulb in our home is perhaps the most simple task in which we can still credit ourselves for performing household maintenance. The amount of thought such an operation receives rarely extends beyond looking for the most efficient pack of bulbs on a store shelf. What happens however, when a municipal utility department has over 100,000 streetlights to change? 

The ideal dusk-to-dawn area light is often defined by lamp efficacy, while lamp efficacy is defined by the operating parameters of human vision. These parameters of human vision are defined by a series of laboratory tests, looking to answer a seemingly simple question – what spectrum of light enables the highest efficacy for vision on a roadway at night?

Depending on where you live, a lighting standards organization such as the CIE (Commission Internationale de l’Eclairage or International Commission on Illumination) or IESNA (Illuminating Engineering Society of North America more often called the IES) has provided your municipality with documents to answer this question. To simplify this exchange of information, a few assumptions have been made; While roadways are illuminated, streetlights themselves are not visible. As a driver neither you nor your fellow road user drives with headlights at night, nor do you ever turn your head or move your gaze throughout the road. If this scenario sounds odd, you aren’t the only one wondering how billions of dollars were invested based on lighting specifications following these assumptions.

During the 1990s, sources of ‘white’ lighting such as metal halide and magnetic induction were raised as alternatives to the industry standard high-pressure sodium lamp. What millions of people simply knew as ‘the yellow streetlights’ were in operation for that exact reason. Yellow is the most efficient portion of visible light for human vision, at least during daytime hours. To understand lighting research, one needs a crash course on human vision:

The human eye contains 2 classes of cells for vision. Cone cells operate in the ‘brighter’, photopic, end of human vision with a peak sensitivity to yellow light. Cone cells are located primarily in the center of the retina, provide sharp ‘focal’ vision and allow us to perceive color. Rod cells operate in the ‘darker’, scotopic, end of human vision with a peak sensitivity to bluish-green light. Rod cells are located primarily in the peripheral region of the retina and provide colorless vision with low resolution. Rod cells require extensive time to generate photopigment in order to achieve sensitivity to light, upwards of 30 minutes for peak sensitivity. It only takes a few seconds, however, for this photopigment to be lost when exposed to bright light. 

Spectral sensitivity of rod cells (left) and that of cone cells (right)
Spectral sensitivity of rod cells (left) and that of cone cells (right)

The primary research behind standards for roadway lighting is the Mesopic Optimization of Visual Efficiency, shortened to M.O.V.E. – mesopic being the range in which both cone and rod cells are theoretically active. MOVE studies consisted of participants in a laboratory performing simulated visual tasks, generally using test stimuli projected onto a background. At one point a driving simulator was utilized, although this also consisted of a simple series of light projections onto a background.

MOVE studies include 3 primary themes worth considering before their application to the real world. Firstly, subjects were given varying amounts of time in a dark room for their eyes to adapt to the luminance level being tested – up to 30 minutes. Secondly, visual performance tests utilized a fixation point and trained subjects to ensure that peripheral vision was being tested when peripheral stimuli was in use. Finally, subjects were never exposed to luminance levels above that of the test surface, up to 10 candela per square meter.

For those of us who don’t live inside a laboratory simulation, we can only dream of driving in the dark for 30 minutes without being subjected to oncoming headlights and other sources of bright light. When exposure to bright light occurs, a variety of biological processes happen within the eyes to adapt to this light. Firstly, the pupil narrows during the Post-Illumination Pupil Response. Shorter wavelengths of light have a much stronger PIPR impact than longer wavelengths of light, as has been observed in a variety of studies. The increased pupil response to blue light is additionally a trigger for physical discomfort.

Comparison of pupil size during and after exposure to different wavelengths of light
Comparison of pupil size during and after exposure to different wavelengths of light.

After PIPR takes place, a series of reactions happen within and between retinal cells. To summarize, the photopigment within rod cells absorbs light and becomes transparent. Should the photopigment become completely transparent, the rod cell is blind until the pigment can regenerate. Cone cells experience an adaptation process as well, albeit with less sensitivity and a much faster recovery time. We often see this as colored ‘splotches’ within our vision, matching the shape of the light source that we viewed. Getting more complex, retinal cells can also receive adaptation ‘instructions’ from their peers. Specifically, cone cells being exposed to bright light results in rod cells across the retina adapting to the light source, resulting in a full-field reduction of light sensitivity.

When the results of the MOVE studies are placed within the context of visual adaptation mechanics, we can make the following conclusions:

  • The theoretical advantage for shorter wavelengths of light in dusk-to-dawn lighting only exists for peripheral vision.
  • This theoretical advantage, mediated by rod cells, exists up until 0.6 cd/m². Roadway luminance targets range between 0.3 and 2.0 cd/m².
  • When driving at night, drivers are exposed to an onslaught of light sources. The luminance of these light sources are several magnitudes greater than the upper limit for rod cell function. Even cone cells are routinely blinded by bright lights, often leading to temporary full-field night blindness.

With these factors in mind, a troubling conclusion for much of the lighting field emerges. In a real-world scenario, there is no evidence that rod cells are able to reliably contribute to vision under standard artificial illumination. Whether you have been informed of a supposed 30% gain in efficiency, or a whopping 70% provided by increased sensitivity to blue light, there was never a foundation behind these numbers. Without reliable vision from rod cells, we are left with our cone cells that perceive yellow light most efficiently. In the second part of this article we will examine real-world driving tests and see how they compare to the laboratory-based models used to select the streetlights outside your home.

Noah Sabatier is a photographer and lighting researcher that is dedicated to advocating for better outdoor lighting. Noah has spent the past 5 years living with a night shift sleep schedule, during this time he realized that the streetlights in his city were far from optimal – and recent changes had only made them worse. He has spent the past 2 years extensively reviewing scientific literature and technical documents alongside others advocating for better lighting. Noah is now working to raise awareness of common misconceptions that lead to bad lighting and the better practices needed to solve this problem.

Reach him at: noahsabatierphoto[at]gmail.com

Works Cited:

Rea M, Bullough J, Freyssinier-Nova J, Bierman A. A proposed unified system of photometry. Lighting Research & Technology. 2004;36(2):85-109
Goodman T, Forbes A, Walkey H, et al. Mesopic visual efficiency IV: a model with relevance to nighttime driving and other applications. Lighting Research & Technology. 2007;39(4):365-392.
Uchida T, Ayama M, Akashi Y, et al. Adaptation luminance simulation for CIE mesopic photometry system implementation. Lighting Research & Technology. 2016;48(1):14-25
Bommel, W. Road Lighting. Fundamentals, Technology and Application; Springer: Berlin/Heidelberg, Germany, 2015. Doi: 10.1007/978-3-319-11466-8
Sharpe LT, Fach CC, Stockman A. The field adaptation of the human rod visual system. J Physiol. 1992 Jan;445:319-43. doi: 10.1113/jphysiol.1992.sp018926. PMID: 1501137; PMCID: PMC1179984.
Lei, Shaobo & Goltz, Herb & Chandrakumar, Mano & Wong, Agnes. (2014). Full-Field Chromatic Pupillometry for the Assessment of the Post-Illumination Pupil Response Driven by Melanopsin-Containing Retinal Ganglion Cells.. Investigative ophthalmology & visual science. 55. 10.1167/iovs.14-14103.

The Streetlight Effect

The Streetlight Effect: Modern Considerations of Early Observations in the Psychology of Outdoor Lighting

By Noah Sabatier

It’s no secret that people don’t enjoy searching for something in the dark. Shadows dance, shapes shift and forms seemingly appear out of nowhere. The Streetlight Effect originated as an early 1900s anecdote in which a drunken man is searching for his keys. A police officer helps him search, resulting in both men spending several minutes under a streetlight. When the officer asks if the drunken man lost his keys under the streetlight he replies “no, this is where the light is”. 


The stakes become much higher while driving, when individuals have mere seconds to see, identify and react to obstacles. This visual ability can be the difference between life and death, as new statistics suggest. 2022 saw a 40-year high of pedestrian deaths in the US, with the trend growing at an alarming rate. Commonly blamed for this rise in pedestrian fatalities is mobile device usage by drivers and the gradual increase in average vehicle size. The statistics paint a different picture however; 

While these statistics don’t reveal the cause of every fatality, we can safely conclude that vehicles do not magically shrink after sunset, nor do mobile devices vanish at night. With such a sharp rise specifically in nighttime fatalities we need to ask; What went wrong with roadway lighting and vehicle headlights?

Research on driver behavior confirms the aforementioned streetlight effect. In 2013 the gaze direction of drivers was examined on a stretch of road during day and night. During daylight hours when sunlight provided illumination of the entire scene, horizontal gaze direction extended beyond either edge of the road. This changed for night driving tests, in which gaze patterns shrunk to match the areas illuminated by headlights and streetlights. There was little to no desire for drivers to spend precious seconds gazing around unlit areas, nor was there a visual stimulus to draw the attention of visual gaze.

 

Further research conducted in 2015 examined the visual behavior of drivers by tracking both their head movements and eye movements to establish gaze direction. 2 roads were utilized for testing, a main route and a residential street both illuminated by streetlights. Collected data revealed that drivers consciously adjust their gaze based on anticipated hazards and vehicle speed. Driving down a main route, the driver’s gaze was narrowed and more time was spent focusing down the road. In the residential area, more time was spent scanning the near side of the road as well as regions where other cars or pedestrians could enter the vehicle’s path.

 

Proper understanding of driver gaze behavior is important, as improper lighting can reduce a driver’s ability to detect targets before they enter the roadway. With the majority of LED fixtures sporting a lower efficacy than the sodium lights they replaced, especially in early transitions, their energy savings relied on illuminating a smaller area to a lower level of luminance. While this is a case-by-case issue for different transitions, common trends can be observed. In many cases the illumination of sidewalks was reduced, both behind and infront of LED streetlights. In fewer cases, dark spots were created between light posts where the LED light was unable to match the illumination area of the previous sodium light fixture. Neither of these are mistakes, as a primary selling feature of LED fixtures is energy savings through tighter illumination patterns.

Understanding both visual psychology and the biological mechanics of the retina is critical to implementing lighting optimally. For an industry and professional field increasingly focused on reducing light pollution, minimizing illuminated areas is often considered a primary goal. Caution must be used however, as research reveals that driver gaze extends beyond the roadway in anticipation of hazards based on the driving scenario.  Should new lighting systems reduce luminance levels on a sidewalk or highway shoulder, drivers may lose their ability to detect hazards before they enter the vehicle’s path. If the goal of visibility and safety is a priority for the lighting field, our gaze must widen to issues of visual system psychology.

Noah Sabatier is a photographer and lighting researcher that is dedicated to advocating for better outdoor lighting. Noah has spent the past 5 years living with a night shift sleep schedule, during this time he realized that the streetlights in his city were far from optimal – and recent changes had only made them worse. He has spent the past 2 years extensively reviewing scientific literature and technical documents alongside others advocating for better lighting. Noah is now working to raise awareness of common misconceptions that lead to bad lighting and the better practices needed to solve this problem.

Works Cited:

Cengiz C, Kotkanen H, Puolakka M, et al. Combined eye-tracking and luminance measurements while driving on a rural road: Towards determining mesopic adaptation luminance. Lighting Research & Technology. 2014;46(6):676-694. doi:10.1177/1477153513503361
Governors Highway Safety Association 2022 Preliminary Data
https://www.ghsa.org/sites/default/files/2023-06/GHSA%20-%20Pedestrian%20Traffic%20Fatalities%20by%20State%2C%202022%20Preliminary%20Data%20%28January-December%29.pdf
Fujiyama, Taku & Childs, Craig & Boampong, Derrick & Tyler, Nick. (2007). How do elderly pedestrians perceive hazards in the street? –
An initial investigation towards development of a pedestrian simulation that incorporates reaction of various pedestrians to environments. Social Research in Transport (SORT) Clearinghouse.
Gibbons, Ronald B.;Meyer, Jason E.;Terry, Travis N.;Bhagavathula, Rajaram;Lewis, Alan;Flanagan, Michael;Connell, Caroline; Evaluation of the Impact of Spectral Power Distribution on Driver Performance; Virginia Tech Transportation Institute, United States. Federal Highway Administration. Office of Safety Research and Development Report Number : FHWA-HRT-15-047

Lighting for the Aging Eye

Lighting for the Aging Eye

By Noah Sabatier

In 2022 nearly 1/3rd of the US population was over the age of 55. Many of us personally know someone who has struggled with vision as they age, these challenges becoming most present during driving. This issue is reflected in research, with older drivers displaying slower reaction times and higher collision rates compared to younger drivers. Driving performance differences become amplified at night, a time in which the aging eye has its greatest impact on visibility.

The most visible aspect of the aging eye is a constricted pupil opening. Physically, the pupil is the first means of adaptation to light. As the eye ages, the maximum diameter of the pupil dwindles to half of its former size at a younger age. Consequently we see a loss in the quantity of light that is able to physically reach the retina.

Once a diminished quantity of light enters the pupil, older individuals face another challenge in visibility. Over time the lens of the eye yellows and takes on a cloudy appearance, modifying the spectral transmittance of the lens. Large quantities of blue light are absorbed by the lens, along with some degree of green light. Yellow, amber and red light is largely unaffected by the yellowed lens. When the impacts of a narrowed pupil diameter and yellowed lens are considered, the loss of light transmission is comparable to a young person wearing sunglasses at night.

Spectral transmission of the eye’s lens for ages 50 and 65, relative to age 25.

Within the Mesopic range of vision, in which both Cone cells and Rod cells are mathematically active, the greatest determining factor of visual performance is luminance levels. In the majority of Mesopic visual performance studies, subjects were in their 20s and 30s. This is important to note, as the goal of Mesopic visual performance studies centers around finding the optimum spectrum for lighting, as well as the minimum luminance level needed to achieve a specified level of visibility. Without taking factors of the aging eye into account, we end up with a Mesopic visual performance model that does not represent people of all ages.

Issues with a lack of accounting for age can be seen in the following scenario; Generic Roadway Lighting Guide lists 0.3 cd/m^2 as the appropriate luminance target for a roadway based on its traffic and speed limit. This number is derived from Mesopic visual performance models, in which 0.3 cd/m^2 correlates to factors of visual performance such as reaction time and the ability to discern object details. While younger individuals will experience the predicted level of visual performance, older individuals are left behind. Unable to physically receive the same quantity of light, their visual performance is below that of the intended target. 

A much greater issue brought about in the era of ‘white’ LEDs is that of a shifting spectral sensitivity with age. When the lens of someone aged 50 is compared to that of someone aged 25, the transmission of deep blue light is reduced by over 50%. Cyan transmittance is reduced by ~25%, while losses for yellow light are around 10% or less. In his technical guide, lighting researcher Wout van Bommel calculated the difference in light transmission for someone aged 25 and someone aged 50 under LEDs. The light from a 4000K ‘neutral’ LED was 11% less visible than light from a 2700K ‘warm’ LED. 

If lighting practitioners wish to provide the best visibility possible to individuals of all ages, several considerations need to be made. Older individuals require a higher luminance level in order to achieve the same visual performance as a younger individual. The difference in required luminance levels becomes greater as the short wavelength content of a light source increases. While increasing luminance targets is a contentious issue, especially among light pollution advocates, there is little question as to the ideal spectrum of light sources for older individuals; The yellow-amber spectrum.  Should we desire to reduce accident rates for the most at-risk class of drivers, implementing lighting optimized for all ages is an easy step to take.

Noah Sabatier is a photographer and lighting researcher that is dedicated to advocating for better outdoor lighting. Noah has spent the past 5 years living with a night shift sleep schedule, during this time he realized that the streetlights in his city were far from optimal – and recent changes had only made them worse. He has spent the past 2 years extensively reviewing scientific literature and technical documents alongside others advocating for better lighting. Noah is now working to raise awareness of common misconceptions that lead to bad lighting and the better practices needed to solve this problem.

Works Cited:

https://www.census.gov/data/tables/2022/demo/age-and-sex/2022-older-population.html
Shadi Doroudgar, Hannah Mae Chuang, Paul J. Perry, Kelan Thomas, Kimberly Bohnert & Joanne Canedo (2017) Driving performance comparing older versus younger drivers, Traffic Injury Prevention, 18:1, 41-46, DOI: 10.1080/15389588.2016.1194980
Walkey H, Orreveteläinen P, Barbur J, et al. Mesopic visual efficiency II: reaction time experiments. Lighting Research & Technology. 2007;39(4):335-354. doi:10.1177/1477153507080920
van de Kraats J, van Norren D. Optical density of the aging human ocular media in the visible and the UV. J Opt Soc Am A Opt Image Sci Vis. 2007 Jul;24(7):1842-57. doi: 10.1364/josaa.24.001842. PMID: 17728807.
Preciado O, Manzano E. Spectral characteristics of road surfaces and eye transmittance: Effects on energy efficiency of road lighting at mesopic levels. Lighting Research & Technology. 2018;50(6):842-861. doi:10.1177/1477153517718227
Vicente, E.G., Arranz, I., Issolio, L. et al. Influence of age and spectral power distribution on mesopic visual sensitivity. Atten Percept Psychophys 81, 504–516 (2019). https://doi.org/10.3758/s13414-018-1616-6
Bommel, W. Road Lighting. Fundamentals, Technology and Application; Springer: Berlin/Heidelberg, Germany, 2015. Doi: 10.1007/978-3-319-11466-8